LiveJournal is a social network journal service with focus on user interactions. As for many other online social networks, predicting potential friendships in the LiveJournal network is a problem of great practical interest. Previous work has shown that graph features extracted from the graph associated with the network are good predictors for friendship links. However, contrary to the intuition, user data (e.g., interests shared by two users) does not always improve the predictions obtained with graph features alone. This could be due to the fact that features constructed from a large number of user declared interests cannot capture the implicit semantic of the interests. To test this hypothesis, we use a clustering approach to build an interest ontology, and explore the ability of the ontology to improve the performance of learning algorithms at predicting friendship links, when interest-based features are used alone or in combination with graph-based features. The results show that ontology-based features can help improve the performance of several machine learning classifiers (in particular, random forest classifiers) at the task of predicting links in the LiveJournal social network.

This paper explores the following question: what kind of statistical guarantees can be given when doing variable selection in high-dimensional models? In particular, we look at the error rates and power of some multi-stage regression methods. In the first stage we fit a set of candidate models. In the second stage we select one model by cross-validation. In the third stage we use hypothesis testing to eliminate some variables. We refer to the first two stages as "screening" and the last stage as "cleaning." We consider three screening methods: the lasso, marginal regression, and forward stepwise regression. Our method gives consistent variable selection under certain conditions.

Motivation: Prediction of ligands for proteins of known 3D structure is important to understand structure-function relationship, predict molecular function, or design new drugs. Results: We explore a new approach for ligand prediction in which binding pockets are represented by atom clouds. Each target pocket is compared to an ensemble of pockets of known ligands. Pockets are aligned in 3D space with further use of convolution kernels between clouds of points. Performance of the new method for ligand prediction is compared to those of other available measures and to docking programs. We discuss two criteria to compare the quality of similarity measures: area under ROC curve (AUC) and classification based scores. We show that the latter is better suited to evaluate the methods with respect to ligand prediction. Our results on existing and new benchmarks indicate that the new method outperforms other approaches, including docking. Availability: The new method is available at http://cbio.ensmp.fr/paris/ Contact: mikhail.zaslavskiy@mines-paristech.fr

The problem of multiple hypothesis testing arises when there are more than one hypothesis to be tested simultaneously for statistical significance. This is a very common situation in many data mining applications. For instance, assessing simultaneously the significance of all frequent itemsets of a single dataset entails a host of hypothesis, one for each itemset. A multiple hypothesis testing method is needed to control the number of false positives (Type I error). Our contribution in this paper is to extend the multiple hypothesis framework to be used with a generic data mining algorithm. We provide a method that provably controls the family-wise error rate (FWER, the probability of at least one false positive) in the strong sense. We evaluate the performance of our solution on both real and generated data. The results show that our method controls the FWER while maintaining the power of the test.

We introduce a symmetry reduction technique for model checking temporal-epistemic properties of multi-agent systems defined in the mainstream interpreted systems framework. The technique, based on counterpart semantics, aims to reduce the set of initial states that need to be considered in a model. We present theoretical results establishing that there are neither false positives nor false negatives in the reduced model. We evaluate the technique by presenting the results of an implementation tested against two well known applications of epistemic logic, the muddy children and the dining cryptographers. The experimental results obtained confirm that the reduction in model checking time can be dramatic, thereby allowing for the verification of hitherto intractable systems.

Mechanical design tools would be considerably more useful if we could interact with them in the way that human designers communicate design ideas to one another, i.e., using crude sketches and informal speech. Those crude sketches frequently contain pen strokes of two different sorts, one type portraying device structure, the other denoting gestures, such as arrows used to indicate motion. We report here on techniques we developed that use information from both sketch and speech to distinguish gesture strokes from non-gestures -- a critical first step in understanding a sketch of a device. We collected and analyzed unconstrained device descriptions, which revealed six common types of gestures. Guided by this knowledge, we developed a classifier that uses both sketch and speech features to distinguish gesture strokes from non-gestures. Experiments with our techniques indicate that the sketch and speech modalities alone produce equivalent classification accuracy, but combining them produces higher accuracy.

Building and using agent-based models is often impractical, in part due to the cost of including expensive subject matter experts (SMEs) in the development process. In this paper, we describe a method for "bootstrapping" model building to lower the cost of overall model development. The models we are interested in here capture dynamic phenomena related to international and subnational conflict. The method of acquiring these models begins with event data drawn from news reports about a conflict region, and infers model characteristics particular to a conflict modeling framework called the Power Structure Toolkit (PSTK). We describe the toolkit and how it has been used prior to this work. We then describe the current problem of modeling conflict and the empirical data available to learn models, and extensions to the PSTK for model generation from this data. We also describe a formative evaluation of the system that compares the performance and costs of models built entirely by an SME against models built with an SME aided by the automated model generation process. Early results indicate at least equivalent prediction rates with significant savings in model generation costs.

Practical data analysis relies on the ability to count observations of objectssuccinctly and efficiently. Unfortunately the space usage of an exact estimator grows with the size of the a priori set from which objects are drawn while the time required to maintain such an estimator grows with the size of the data set. We present static and on-line approximation schemes that avoid these limitations when approximate frequency estimates are acceptable. Our Log-Frequency Sketch extends the approximate counting algorithm of Morris [Morris1978] to estimate frequencies with bounded relative error via a single pass over a data set. It uses constant space per object when the frequencies follow a power law and can be maintained in constant time per observation. We give an (epsilon, delta)-approximation scheme which we verify empirically on a large natural language data set where, for instance, 95 percent of frequencies are estimated with relative error less than 0.25 using fewer than 11 bits per object in the static case and 15 bits per object on-line.

Detecting abnormal activities from sensor readings is an important research problem in activity recognition. A number of different algorithms have been proposed in the past to tackle this problem. Many of the previous state-based approaches suffer from the problem of failing to decide the appropriate number of states, which are difficult to find through a trial and-error approach, in real-world applications. In this paper, we propose an accurate and flexible framework for abnormal activity recognition from sensor readings that involves less human tuning of model parameters. Our approach first applies a Hierarchical Dirichlet Process Hidden Markov Model (HDP-HMM), which supports an infinite number of states, to automatically find an appropriate number of states. We incorporate a Fisher Kernel into the One-Class Support Vector Machine (OCSVM) model to filter out the activities that are likely to be normal. Finally, we derive an abnormal activity model from the normal activity models to reduce false positive rate in an unsupervised manner. Our main contribution is that our proposed HDP-HMM models can decide the appropriate number of states automatically, and that by incorporating a Fisher Kernel into the OCSVM model, we can combine the advantages from generative model and discriminative model. We demonstrate the effectiveness of our approach by using several real-world datasets to test our algorithm’s performance.

Bootstrap Voting Experts (BVE) is an extension to the Voting Experts algorithm for unsupervised chunking of sequences. BVE generates a series of segmentations, each of which incorporates knowledge gained from the previous segmentation. We show that this method of bootstrapping improves the performance of Voting Experts in a variety of unsupervised word segmentation scenarios, and generally improves both precision and recall of the algorithm. We also show that Minimum Description Length (MDL) can be used to choose nearly optimal parameters for Voting Experts in an unsupervised manner.